Internalizing and externalizing disorders occur together at high rates, but few studies have explicitly attempted to explain this pattern. The current application seeks to investigate the co-occurrence of aggression and anxiety/depression using data from multiple reporters, and to use measures of temperament and parenting to predict different patterns of symptom development, including co-occurrence, using modern statistical methods. The participants in this study were 214 children (many of whom were identified as being at risk for clinical disorder using a prescreening measure) who were assessed every two years for eight years, starting when they were age 4.5 to 8. Parents and teachers completed questionnaires evaluating children's mental health problems at all five assessments, and adolescents completed self-report measures of mental health symptoms at the final two assessments. Temperament was assessed using mothers' reports at the first study visit, and parenting behavior will be coded from video recordings of a timed parent-child puzzle task, also from the first study visit. Often different respondents disagree in their assessment of mental health issues. The first goal of this study is to examine the degree to which mothers, fathers, teachers, and adolescents are similar in their assessment of aggression and anxiety/depression, and to investigate the nature of any systematic differences between these reporters. The second major goal of this investigation is to classify children into different trajectories of symptom development (e.g., co-occurring aggression and anxiety/depression; high aggression; high anxiety/depression; low symptom), and to differentially predict membership in these different classes using measures of parenting and temperament. It is expected that this study will provide valuable information about the development of (and risk factors for) different patterns of mental health symptomatology.